This project (fMRI) will use fMRI techniques to examine the functional integrity of the neuroanatomic networks subserving episodic memory and executive function, abilities that demonstrate significant impairment in prodromal AD. A series of fMRI experiments are planned to evaluate older controls and non-demented individuals with varying levels of cognitive impairment, who are at risk for progressing to meet clinical criteria for AD. Both block design and event-related paradigms will be employed. Experiments 1-3 will employ a face-name association memory paradigm, which had shown reliable activation in a specific functional network, that includes the anterior hippocampus, fusiform and prefrontal cortices. Experiment 4 is focused on the functional network activated during an executive function paradigm, a set-switching and interference task that has shown reliable activation in the anterior cingulate. By examining patterns of fMRI activation during episodic memory and executive function, we hope to elucidate the neural underpinnings of the cognitive dysfunction that occurs in prodromal D. We will test the hypothesis that alterations in the fMRI activation in specific regions can differentiate individuals with varying levels of cognitive impairment, and determine if longitudinal change in activation is associated with change in level of cognitive impairment. We will also examine the relationship of fMRI data to variables from other cores and projects.